QUIETT: Query-Independent Table Transformation for Robust Reasoning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912921046482944 |
|---|---|
| author | Najpande, Gaurav Kumar, Tampu Ravi Choudhury, Manan Roy Valeti, Neha Fu, Yanjie Gupta, Vivek |
| author_facet | Najpande, Gaurav Kumar, Tampu Ravi Choudhury, Manan Roy Valeti, Neha Fu, Yanjie Gupta, Vivek |
| contents | Real-world tables often exhibit irregular schemas, heterogeneous value formats, and implicit relational structure, which degrade the reliability of downstream table reasoning and question answering. Most existing approaches address these issues in a query-dependent manner, entangling table cleanup with reasoning and thus limiting generalization. We introduce QuIeTT, a query-independent table transformation framework that preprocesses raw tables into a single SQL-ready canonical representation before any test-time queries are observed. QuIeTT performs lossless schema and value normalization, exposes implicit relations, and preserves full provenance via raw table snapshots. By decoupling table transformation from reasoning, QuIeTT enables cleaner, more reliable, and highly efficient querying without modifying downstream models. Experiments on four benchmarks, WikiTQ, HiTab, NQ-Table, and SequentialQA show consistent gains across models and reasoning paradigms, with particularly strong improvements on a challenge set of structurally diverse, unseen questions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20017 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | QUIETT: Query-Independent Table Transformation for Robust Reasoning Najpande, Gaurav Kumar, Tampu Ravi Choudhury, Manan Roy Valeti, Neha Fu, Yanjie Gupta, Vivek Computation and Language Real-world tables often exhibit irregular schemas, heterogeneous value formats, and implicit relational structure, which degrade the reliability of downstream table reasoning and question answering. Most existing approaches address these issues in a query-dependent manner, entangling table cleanup with reasoning and thus limiting generalization. We introduce QuIeTT, a query-independent table transformation framework that preprocesses raw tables into a single SQL-ready canonical representation before any test-time queries are observed. QuIeTT performs lossless schema and value normalization, exposes implicit relations, and preserves full provenance via raw table snapshots. By decoupling table transformation from reasoning, QuIeTT enables cleaner, more reliable, and highly efficient querying without modifying downstream models. Experiments on four benchmarks, WikiTQ, HiTab, NQ-Table, and SequentialQA show consistent gains across models and reasoning paradigms, with particularly strong improvements on a challenge set of structurally diverse, unseen questions. |
| title | QUIETT: Query-Independent Table Transformation for Robust Reasoning |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.20017 |